The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process the content of arbitrary code files within a user's convex/ directory. This creates a surface for indirect prompt injection if those source files contain malicious instructions disguised as code or comments. Ingestion points: Processes all source code files found within the convex/ directory. Boundary markers: No explicit delimiters or instructions to ignore embedded prompts are provided in the skill logic. Capability inventory: The skill is configured to read files and output a formatted report to the user. Sanitization: No sanitization or filtering of the ingested code content is specified before the agent processes it.

convex-reviewer